Real Time Surveillance of Influenza in Ambulatory Primary Care

BACKGROUND:    Delays in reporting are inherent in syndromic, mechanistic and laboratory surveillance of influenza.  Consequently, public health acquisition of information and response to influenza outbreaks is typically offset by two-to-three weeks.  The feasibility of creating and operating a statewide array of rapid influenza detection test (RIDT) analyzers that report results in real time was assessed.  The analyzers use wireless connectivity to communicate results to a cloud-based server. 

METHODS:   The primary outcome was the feasibility of a real-time primary care influenza surveillance network, including the time to recruit, engage, and train practices and for installation and implementation of analyzers.  Secondary objectives included the assessment of performance in the initial influenza season for detection of influenza as compared to existing surveillance programs.  For our array of clinical surveillance sites, we used the Quidel Sofia Influenza A+B FIA—a RIDT with wireless connectivity.  At least two analyzers were deployed in each of five public health regions of Wisconsin, starting in October 2013.  Twenty primary care practices located in urban, suburban and rural locations (including 16 Wisconsin Research and Education Network sites) were recruited and participated. 

RESULTS:   Initial IRB exemption occurred on 6/25/2013.  University financial clearances were attained on 8/20/2013.  Practices were contacted and recruited thereafter with full complement attained by 11/20/2013.  Installations of Sofia analyzers were completed by 12/18/2013.  Data were accumulated as soon as sites were activated and reporting from 16 sites (80%) was fully operationalized by 12/31/2013.  This time course allowed for any local site IRB applications and for inserting and applying new technology into practices.  Data were aggregated and analyzed on a daily and weekly basis by site, public health region, and for the Wisconsin composite.  The system identified the onset of the 2013-2014 seasonal influenza outbreak extremely early. 

CONCLUSIONS:   Effortless, real time reporting of RIDT results was achievable over a very short time frame in practices using RIDT coupled with immediate wireless transmission of results.  Such reporting eliminated the need for clinicians or laboratorians to identify time to aggregate and transmit information.  This approach is a reasonable model for public health surveillance for any pathogen identifiable by clinic-based technology.